Research On Cache Based Database Index

A cache is a smaller,faster memory,closer to the CPU,which stores copies of the data from frequently used main memory locations.When the CPU needs to read from or write to a location in main memory,it first checks whether a copy of that data is in the cache.If so,the processor immediately reads from or writes to the cache,which is much faster than reading from or writing to main memory.If not,CPU has to wait until cache fetch the data from memory,which means a CPU stall.the time taken to fetch one cache line from memory matters because the CPU will run out of things to do while waiting for the cache line.Analysis of the components of the stall time indicates that on the average,half the execution time is spent in stalls.In all cases,90% of the stalls are due to second-level cache data misses and first-level instruction cache misses.So it's import to improve cache-conscious ability of database index.(1)Increment of data indicates index will consumer much more memory space and will certainly cause losts of swap between cache and memory.So I adopted compression to index and propose DCST-tree.DCST-tree reduce space consumption of index by delta-encoding technique.Meanwhile,it uses two-level node layout to accelerate search process in a node.(2)HBase can not search column information efficiently,it has to scan the whole user data.Lots of useless data will pollute cache and consume a lot of cache space.So I propose SIP index.SIP index can narrow the range of scan.It can also filter useless data in case not to comsumer cache space.(3)The thesis designs and implements an energy information query system based on the framework of Spark Streaming,HBase and Kafka.The System provide ability for big data query.The design of the system,architecture,implementation of various modules to do a detailed introduction.